transcribe-filing

transcribe-filing is a skill for Claude Code, Codex from vadim-filimonov/openrater. It costs 102 tokens per session (3,073 once invoked), scanned A, original, Apache-2.0.

A document-to-rating workflow that turns an insurance rate filing or internal rating manual into an OpenRater workbook and executable rating plan. OpenRater is a calculator that validates the workbook and calculates rates.

In plain words
What is it for?
Use it to digitize a filing, validate the resulting workbook, build a rating plan, check it against test cases, and rate risks through OpenRater.
Why use it?
It removes the need to manually turn rules in PDFs or manuals into rating logic or guess missing factors. It also pauses for human review when validation errors, gaps, or failed test cases appear.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: positional $N argument; mentions Claude Code.

Good fit Use it to digitize a filing, validate the resulting workbook, build a rating plan, check it against test cases, and rate risks through OpenRater.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/vadim-filimonov/openrater/transcribe-filing
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add vadim-filimonov/openrater --skill transcribe-filing
Clone the repo
git clone --depth 1 https://github.com/vadim-filimonov/openrater

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for transcribe-filing

README.md
[![agentmods](https://agentmods.dev/badge/skills/vadim-filimonov/openrater/transcribe-filing/github.svg)](https://agentmods.dev/skills/vadim-filimonov/openrater/transcribe-filing)
Your own site
<a href="https://agentmods.dev/skills/vadim-filimonov/openrater/transcribe-filing"><img src="https://agentmods.dev/badge/skills/vadim-filimonov/openrater/transcribe-filing/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for transcribe-filing

Your own site · 80×15
<a href="https://agentmods.dev/skills/vadim-filimonov/openrater/transcribe-filing"><img src="https://agentmods.dev/badge/skills/vadim-filimonov/openrater/transcribe-filing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 102 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,073 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00102 $0.03073
Opus 5 $0.00051 $0.01537
Sonnet 5 $0.00020 $0.00615
Haiku 4.5 $0.00010 $0.00307

Measured 11d ago against content hash 24d9d402a4f5, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

transcribe-filing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/transcribe-filing/SKILL.md · 231 lines

How it starts

The opening of the file, as written. The whole thing — 231 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Transcribe a filing into an executable rating plan

You are the transcriber. OpenRater is the calculator. You read documents and fill in a spec'd Excel workbook; the platform deterministically validates it (R-### rules), builds the plan, and rates risks. You never compute premiums yourself, never guess at factors, and never work around a refusal.

Non-negotiable rules

  1. Review stops. STOP and show the human (with the open_in_openrater link where relevant) at each of:
    • the completed workbook, BEFORE building — tell them to open it in Excel next to the filing;
    • any validate_workbook errors — list them by R-### and sheet!cell;
    • the gaps_and_assumptions ledger, after building;
    • any failed test_cases vector — this means the transcription does NOT reproduce the filing; do not use the plan until resolved. Never continue past a stop without an explicit go-ahead.
  2. Bulk data stays out of the chat. Books and CSVs travel as file paths; report counts and summaries only.
  3. Record gaps; never approximate. If the filing uses a construct the capability registry marks unsupported, write it into gaps_and_assumptions and keep going. A visible gap beats a wrong number nobody can see.
  4. The user supplies the documents. Never fetch from SERFF or any filing portal yourself — no scraping, ever.
  5. Filed rates govern. Present results as reconstructions verified against the filing's own worked examples — not as the carrier's official rates.

First run — the three doors (offer once, then never again)

On the user's FIRST substantive OpenRater message of a conversation — when nothing in the conversation shows they've used it before — orient with ONE short menu, then follow their pick. Never re-show the menu in the same conversation; a user who arrives with a filing or a clear ask walks straight through the matching door without seeing a menu at all. (Same behavior in Claude Desktop and Claude Code.)

Read the full file on GitHub · 231 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 11d ago First seen · 231 lines · 102 tokens per session scan A 24d9d402a4f5

Subscribe to this mod's changes

transcribe-filing is a skill published in the GitHub repository vadim-filimonov/openrater (0 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 102 tokens to every session and 3,073 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.